Bootstrap Aggregation and Cross‐Validation Methods to Reduce Overfitting in Reservoir Control Policy Search. Issue 8 (19th August 2020)

• Record Type: Journal Article
• Authors: Brodeur, Zachary P.; Herman, Jonathan D.; Steinschneider, Scott
• Is Part Of: Water resources research. Volume 56:Issue 8(2020)
• Page Start: n/a
• Page End: n/a
• Publication Date: 2020-08-19
• Subjects: policy search -- machine learning -- paleohydrology -- validation -- reservoir operations -- water resources; Hydrology -- Periodicals; 333.91
• DOI: 10.1029/2020WR027184
• Languages: English
• ISSNs: 0043-1397

Abstract

Abstract: Policy search methods provide a heuristic mapping between observations and decisions and have been widely used in reservoir control studies. However, recent studies have observed a tendency for policy search methods to overfit to the hydrologic data used in training, particularly the sequence of flood and drought events. This technical note develops an extension of bootstrap aggregation (bagging) and cross‐validation techniques, inspired by the machine learning literature, to improve reservoir control policy performance on out‐of‐sample hydrological sequences. We explore these methods using a case study of Folsom Reservoir, California, using control policies structured as binary trees, and streamflow resampling based on the paleo‐inflow record. Results show that calibration‐validation strategies for policy selection coupled with certain ensemble aggregation methods can improve out‐of‐sample performance in water supply and flood risk objectives over baseline performance given fixed computational costs. Our findings highlight the potential to improve policy search methodologies by leveraging these well‐established model training strategies from machine learning. Key Points: We apply machine learning techniques of bootstrap aggregation (bagging) and cross‐validation to improve reservoir control policy search Block bootstrapping of historic hydrology based on paleo‐inflows can efficiently generate calibration‐validation‐testing data Policy selection according to validation performance on bootstrapped data leads to the greatest improvement in out‐of‐sample performance